Design and Select Pumps

Report 8, 1991
DESIGN AND SEIECITON OF DEEP

WELL PUMPS FOR GEOTHERMAL WEllS

Udom Kunaruk,

UNU Geothermal Training Programme, Orkustofnun - National Energy Authority, Grensasvegur 9,

108 Reykjavik,

ICELAND

Permanent Address:

Electricity Generating Authority of Thailand, Thermal Power Engineering Department, Nonthaburi 11000,

THAILAND

3 ABSTRACT

Deep well pumps are necessary in order to get geothermal fluids to the surface for all purposes. Well testing is required for optimum pump design and selection. Well testing involves collection of data and analysis to determine the well loss coefficient. A systematic calculation procedure is presented. The engineering equations used in detailed pump calculation are identified and explained in a case study. The study shows how seasonal and gradual variations of the water table influence the design of the pumps and selection of the optimum discharge. The proper selection of deep well pumps is important for trouble free operation and optimum production from the well. The relative movement of impellers is a critical parameter that can be calculated; the results are used to make sure that the adjustment of impeller to the bowl housing is in the best position. The proper procedures for installing and pulling a deep well pump are given in detail.

4

TABLE OF CONTENTS

Page

ABSTRACf 3

TABLE OF CONTEN'TS 4

LIST OF FIGURES 5

LIST OF TABLES 5

1. INTRODUCfION 6

2. GEOTHERMAL WELL CHARACTERISTICS 7

2.1 General 7

2.2 Well testing 7

2.2.1 Analytical model used in this study 8

2.2.2 Pumping tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 9

2.2.3 Air-lift tests 11

2.2.4 Injection tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 13

2.3 Downhole pumps for geothermal wells 14

3. DEEP WELLUNESHAFT PUMPS 15

3.1 General..................................................... 15

3.2 Selection of a deep well pump 15

3.2.1 Borehole 16

3.2.2 Pump performance curve 17

3.2.3 Pump pressure break down 17

3.2.4 The minimum submergence of bowl unit 18

3.2.5 The minimum length required of column 19

3.3 Adjustment of the impellers 19

3.3.1 Elongation of shaft due to hydraulic thrust 19

3.3.2 Elongation of the column due to hydraulic thrust 19

3.3.3 Column shortening due to internal pressure 21

3.3.4 Required size of the motor 22

3.4 Material selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 22

3.5 Maintenance 25

3.6 Repairs 25

3.7 Engineering information 26

4. PUMP DESIGN AND SELECTION 30

5. INSTALLATION OF GEOlHERMAL DEEP WELL LINESHAFT PUMPS 41

5.1 General 41

5.2 Installation................................................... 41

6. CONCLUSIONS..................................................... 47

ACKNOWLEDGEMENTS : 48

NOMENCLATURE ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 49

REFERENCES. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 51

5

UST OF FlGURFS

Page

1. Step-drawdown test (best fit by polynomial) 8

2. Step-drawdown test (A.HJQ versus Q) .........•...•..•........•..•....... 8

3. Typical pumping test procedure 10

4. Typical arrangement of equipment for a well test ••..••.••.........••••••.• to

5. Common terms applied to air-lift pumping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 11

6. Air required for pumping . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 12

7. Approximate percent pumping submergence for optimum air-lift efficiency 12

8. Basic components - enclosed lineshaft pump. . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 15

9. Semi-open and enclosed type impellers. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 15

to. Multi pump performance CUlVes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 16

11. SIKH 2900 rpm pump performance curve 17

12. Diagram of definitions 20

13. Weight of shaft with length 24

14. Total weight of deep well string ., , . . . . . . . . . . . . . . . . . . . .. 24

15. Checking shaft for straightness , , 25

16. Preliminary pump calculation , 31

17. Detailed pump calculations , 32

18. Seasonal variation of water table 33

19. Gradual variation of water table , . . . . . . .. 33

20. Plot showing flow rate versus water table 36

21. Shaft elongation , , . .. 42

22. Standard enclosing tube and lines haft projection. . . . . . . . . . . . . . . . . . . . . . . . . .. 43

23. Installation of column assembly .,., , 43

24. Discharge head assemblies 44

25. Water lubrication assembly , 45

26. Hollow shaft driver clutch 45

27. Air line installation and use .,........................................ 46

LIST OF TABLES

1. Recommended pipe sizes for air-lift pumping .. . . . . . . . . . . . . . . . . . . . . . . . . . .. 11

2. Recommended flow ranges for vertical turbine pumps 16

3. Coefficient of pump performance curves 17

4. Sizes of column, coupling and enclosing tube ,............... 23

5. Sizes of bearing stock material required for fabrication . . . . . . . . . . . . . . . . . . . . .. 23

6. Calculated weight of components 24

7. Total weight of components in kg/unit 24

8. Deep well pump material nominal size and maximum allowable wear 26

9. Limitations of standard bowl assemblies 27

to. Mechanical friction in lineshafts . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 28

11. Shaft horsepower ratings , 28

12. Column friction loss ... . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 29

13. Summation of calculated parameters . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .. 39

6

1. INTRODUCI10N

The author of this report was awarded a United Nations University (UNU) fellowship to attend the 1991 Geothermal Training Programme held at the National Energy Authority (Orkustofnun) of Iceland, which lasted from April 23 to October 23, 1991. The training programme started with an introductory lecture course for five weeks. The topics of the lectures related to geothermal development i.e. geothermal energy in the world, planning of geothermal projects, geological exploration, borehole geology and alteration mineralogy, hydrology and hydrogeology, geophysical exploration, chemistry of thermal fluids, environment and process utilization, borehole geophysics (logging), well testing, reservoir engineering, drilling technology, well design, low and high temperature utilization and computer applications.

During the second part of the training programme, UNU fellows went on a field excursion from July 10 to July 17, 1991. The main geothermal fields of Iceland were visited, both low and high temperature fields. During the excursion, the fellows received lectures and seminars in the respective areas on geological and geophysical exploration, utilization and the stages of development in each geothermal project.

In the last part of the specialized training programme, the author received training in deep well pump design and attended special lectures in reservoir engineering, well testing, and geochemistry for three weeks. This report will describe the design and selection of deep well pumps and the interpretation of well test data.

Deep well pumps are used to pump geothermal fluids to the surface and onward to the user. As most low temperature geothermal reservoirs are non-artesian, the wells will not produce without pumping. In artesian flow reservoirs, pumps may be installed to increase the flowrate, and in cases where the temperature is above boiling or in wells with a high gas concentration, the pumps may be used to pressurize the water so that it will not boil nor release the gas. The latter case is to be found at Fang in Northern Thailand where two artesian wells are producing flow water for a 300 kW binary turbine. The water temperature is 130°C and the flowrate approximately 22 1/s from the 6" diameter 150 m deep wells. The well fluid (steam and water) is piped to a separator at 1 bar pressure where the steam is vented to the atmosphere, but the hot water is pumped to the power plant. These wells suffer from calcite scaling in surface pipes and the top 10 m of the well. The scales have to be removed by a mechanical cleaning device suspended on a cable every 2 weeks. This procedure takes approximately two hours, and the well output declines over the production period down to approximately 15 1/s. One method that would improve the operation of the plant would be to install a deep well pump. The benefits would be:

1. Greater generating capacity and no reduction in output due to well scaling.

2. Increased production from each well, by lowering of water level.

3. Higher wellhead temperature.

4. No loss of steam to the atmosphere; Better energy recovery.

5. Scaling of calcite will be reduced, but the introduction of a scale inhibitor into the well below the pump to protect the pump will have to be investigated.

In this study, design data for a geothermal well in Iceland with a similar temperature was used, as the required pump test data is not available for the Fang wel1. The calculation procedure developed can, however, be used once the required data has been collected.

7

2 GEOTIIERMAL WELL CHARACfERISTICS

21 General

The main objective of this study is to evaluate the feasibility of using deep well pumps to overcome the calcite scaling problem in low and medium temperature geothermal wells. Evaluation of the site's specific conditions is an important part of designing such a pump system. Several tests and measurements have to be made to establish the temperature of the well, the yield-drawdown relationship and the partial pressure of carbon dioxide. This information is used to make tbe design calculations and in the selection of tbe proper pump system. This chapter will describe how the well and reservoir characteristics are obtained and interpreted.

22 Well testing

Before a downhole pump can be selected for pumping fluid from a geothermal well, we must have an idea of the well and reservoir characteristics. Well testing methods have been used for decades to evaluate geothermal wells and reservoirs (Kjaran and Eliasson, 1983). There are several methods of testing such as pumping tests, injection tests, and recovery tests. Well tests will give information on the hydrological condition of the well and reservoir system. Future prediction on well discharge and drawdown in the reservoir can also be made.

During the well test, water is either discharged from or injected into the well. This will cause pressure changes, which are either monitored in the production well itself or in an observation well.

The parameters that are generally monitored or estimated during well testing are

: time, since well test started (s);

: discharge from or injection into the well being tested (Vs or kg/s); : water level in the well (m);

: radius of discharge well (m);

: distance between the discharge well and the observation well (m); : dynamic viscosity of the reservoir fluid (Pa.s);

: density of reservoir fluid (kg/m''),

The main parameters obtained in analyzing well test data are

Transmissivity = khh: (m3/pas) Storativity = ¢Ch (m/Pa)

where

k : permeability of the reservoir (mz)

cp : porosity of the reservoir rocks

c : compressibility (Pa-l)

h : reservoir thickness (m)

The initial water level, laminar drawdown coefficient and turbulent drawdown coefficient, which are necessary for deep well pump design are determined by use of J acob's and Rorabaugh's method or from Theis solution to the diffusivity equation describing fluid flow in a porous media.

~ JHD HS~ 9OCIO UK

~ L:.t:J 91.10.0066 T ~

GO O.."".,..,..-rr1rTO""TTTT-nZn'"O TTTT-rTTn"30""'" """'-TT'---'4--r'O...,..,.....TTTTSO When drawdown is mostly controlled by well

fLOWRATE Q(l/s) losses (Figure 1), the above equation can be simplified to

This approach is useful for interpreting a step-test

I when the well is pumped at several different rates

r.r:l JH"D HSO 9000 UK d h di d d ed Th

L1:.J 9'.10.0f;67T J an t e correspon 109 raw own measur . . e

o 0 -t-r-r-r-r-r-r-r-r-r-r-r-r-r-r-r """ r-r- r-t- I -r-r- I r-r-r+r-r-e-r-r-r-r-r-r- ..,..-fTTTl"' I ratio .1H/Q can then be plotted against Q which

1 0 fL~~RATE ~(I/ s) 40 50 should yield in a straight line. The coefficients B

and C can be determined, as shown in Figure 2. The linear part of Equation 1 corresponds to the laminar drawdown or the drawdown which occurs in the reservoir due to the well discharge. By idealizing the reservoir and boundary conditions, an equation can be derived which describes the flow of fluid in the reservoir. The resulting equation, known as the diffusivity equation, is obtained by defining and combining Darcy's law, the law of conservation of mass and an equation of state. An approximating solution to the diffusivity equation for an infinite acting reservoir is sometimes called the Theis solution. It's logarithmic approximation can be written as (Todd, 1980):

~H 00337 Q'

50

\

FIGURE 1: Step-drawdown test, (best fit by polynomial)

1.6 ...---------------,

1.2

B 0.155+0B

IIH : BQ+CQ'

IIH/Q : BTCQ

. O.155408+0.0263169Q

(I)

::-

'-'

<,

g08

Cf

<,

:r::

<l

04

/

: C ; (IIH/Q)/Q : 0.0263169

a

F1GURE 2: Step-drawdown test, MfIQ versus Q

8

221 Analytical model used in this study

The water level in a pumping well can be described by the following equation:

H; A+BQ+CQ2

(1)

where

H A B

: water Level (m)

: initial water level (m)

: laminar drawdown coefficient

(m/(Vs)

: turbulent drawdown coefficient (m/(Vs)2)

: well discharge (l/s)

c

Q

(2)

where

.1H : well drawdown (difference between initial and measured water level) (m)

Equation 1 can be simplified for graphical solution often referred to as Jacob's and Rorabaugh's method, by dividing it by Q:

IlH

- = B+CQ

Q

(3)

9

"H 2.30Ql 2.2sn 1 2.2Sn

lJ. '" -- og-- '" m og--

4lt'T r2S 72 S

(4)

Plotting the drawdown, MI, as a function of the logarithm of time, should yield a straight line of slope m. This line can be used to determine the transmissivity of the reservoir,

If we read the change in water level (.dHJO) during one log-cycle from such a graph, then we have (Todd, 1980)

IlH '" 2.30Q· "" m

10 41tT

(5)

which can be solved for the transmissivity coefficient T.

Furthermore, if we extrapolate the semilog straight line to the time to where .dH = 0, Equation 4 can be rearranged to give the following equation, allowing the storativity coefficient (S) to be calculated:

(6)

This approach is generally called the Cooper - Jacob method of solution (Todd, 1980).

As mentioned before, Equation 4 corresponds to the laminar drawdown for transient case in Equation 1. Turbulent drawdown is then equal to total drawdown minus the laminar drawdown,

222 Pumping tests

For a pump test to work correctly and efficiently several criteria need to be fulfilled. The main four are to assure the entry of geothermal fluid into the well and facilitate the operation of pumping equipment, the collection of data from within the well and the measurement of pumped discharge.

The pumping tests are normally carried out to assess the hydraulic behaviour of a well and so determine its usefulness as a SOurce of water, predict its performance under different pumping regimes, select the most suitable pumping plant for long term use and give some estimate of pumping costs. Also, to determine the hydraulic properties of the aquifer or aquifers which yield water to the well. These properties include the transmissivity and related hydraulic conductivities, storage coefficient, laminar pressure drop and turbulent pressure drop, type, and distance of any hydraulic boundaries. Furthermore, it may be necessary to determine the effect of pumping upon neighbouring wells. The pumping tests may be performed by the use of several methods, either with downhole pumping or with an air-lift.

The water level changes in the well and observation wells are measured as a function of time. Each step of the pumping test should be of similar duration. Furthermore, the flow rate and the wellhead temperature are measured. The main phases of a pumping test are shown in Figure 3. Pumping equipment and instruments required to collect data are shown in Figure 4.

10

r:-r:::l JHD HS~ 9000 UK L..:.LJ 91.10.0568 T

Planing

SIan

Time

Fil'li!l.h

Objective I

Study or all available I im!onnalion.

l Pre - [~~ planning

Determine likely I Seleet

dischatge equipment

Objectives and 11'1 i lUll 8.ppt3isaJ

Observations du.ring !0;1.5

Observaucns

Test supervision

constant

d ~i-Cha,rg-e, lo;:!jt

Variable

d lsc h.arge ~e!l\

'Supervision

Special le$($

Presentation or1 Analysis and

da ra I interpretation

Conclusion

FIGURE 3: Typical pumping test procedure (British Standards Institution, 1983)

r7T:I JHO H S p 9000 UK CJ:::...J 91.1 0.066$ T

Pressure gauge Air valve

Sample tap

Power resource

I

HtffiE:~~=- Column pipes

I

la'3::;::c=-- Casing

Aquifer

Main cable

I 1 ~ I I
~F

!

I

~
F
~

.....

t:

I I I r Water level

Foot valve

Pump

Motor

FIGURE 4: Typical arrangement of equipment for a well test (British Standards Institution, 1983)

2.2.3 Air-lift tests

An air-lift test is one pumping test method. The principle of air-lift testing is to use air for lifting the water to the surface by decreasing its density in the welL Compressed air is injected through the air line and through the perforated foot piece on its lower end (Figure 5).

11

r;-r::::1 JHD H$p 9000 UK L....tJ 91 10.0670 T

Plug

Discharge level

Total

'I

starting lift

Air-lift has been found to produce the best results when the submergence ratio of the air line is about 60 percent. Both air pressure and air volume are important in initiating and maintaining an air-lift pumping operation. A useful rule of thumb for determining the proper compressor capacity for air-lift pumping is to provide about 0.007 m3/s (3/4 cfm) of air for each 1 Us (1 gpm) of water at the anticipated pumping rate. To calculate pumping submergence, the length of air line below the water level is divided by the total length of air line. Good results can be obtained by a skilful operator while pumping with a pumping submergence as low as 30% The air line should not be placed all the way to the bottom of the well when pumping begins, unless required for proper submergence,

because the air must then overcome an unnecessarily high pressure head. If steady flow must be maintained, the air volume requirements will usually be greater than those given in Figure 6. For deep wells with low static water level, the actual volume of air required may be two or three times the volume shown in Figure 6 to maintain steady flow. Table 1 gives recommended sizes of educator pipe and air line for air-lift pumping.

'I

'I "

Casing Static

"

"

, I

.: water level

I,

Water

I,

" "I "

drawdown Pumping (.6.H)

... water level

II

I, "

Educator

pipe Pumping

submergence

(h)

-,._

-,~ ,.' ,

(A)

~Ai'I'

-_ r Joe

AGURE 5: Common terms applied to air-lift pumping (Driscoll, 1986)

TABLE 1: Recommended pipe sizes for air-lift pumping (Driscoll, 1986)

Size of educator pipe (or
Pumping Rate Size o[ well casing if educator casing if educator is not Minimum side of air
pipe is used used) line
gpm Us in mm in mm in mm
30-60 2-4 4 102 or more 2 51 0.50 3
60·80 4·5 5 127 or more 3 76 1 25
80·100 5-6 6 152 or more 3.50 89 1 25
100·150 6 + 10 6 152 or more 4 102 1.25 32
150 - 250 10 - 16 8 203 or more 5 127 1.50 38
250·400 16 - 25 8 203 or more 6 152 2 51
400·700 25 - 44 10 254 or more 8 203 3 64
700 . 1,000 44 - 63 12 305 or more 10 254 3 64
1,000 - 1,500 63·9S 16 406 or more 12 305 4 102 450 3 -rn-J~-o- ~~~~-~~- - - - ----- ':' .------ -----'-r--- -- - '--;;:-':

~4251~ ~ ""'O.0671r ._. __ , •••• ~------./-3~--:

E ;:::

~ 4_00 --------··--·-i---·--·-·-----:'---···------~-~--· ·----3~---:

n:: 3.75 .. -------- .. -.~------- ... ----:----- ... ------: -- .. ---------:

~ 3_50 .•. --------- •. ~------- .•. ----~------ .. ----: -----.--------:

~ 275 0..

~ 2.50

~ ::::-c;c>t:,

g 1.75 .---------.,.~ -~- -- ... ~ -----~--.~----------~.----------- .. :

a::: :00::

0:: 1.50 -- ... ------- .~--- ----- .---;------ .. ------(-.-----------!

:;:;: 1.25

_______ • _ _ _ _ ~ __ " ... _. , _~ •• ft. __ :.. • ~ • :

--- 'Y/f-~~' ---; _:~_b_':'_..r_g_e_n:~:- !'.e.r.~~~:_ - ----i

·--~tCe::::-------;-.--------.-..-~------"" .. ----! -·-~-~?·-~-------·------r--··---------·~----------··--:

------·-------,····------··--r-----· ... ··----- .. ·····----------.,

: : : :

0.00 +-r--o-,---r-r-r-r--.--rT-,-,-,-TTT,.,..,---i---r-rTT",...,..-i---,---,--,----;~

400

~ 100 0::

fr_. 0.75

050 0.25

o

100 200 300

, TOTAL PUMPING LIFT (m)

FIGURE 6: Air required for pumping (Driscoll, 1986)

80 r;t'::j - 'HO HS;' 0000 UK - --- - - - - -- -f""- -- - - - - - --:"".

~91_1Q_OEin'T I ;

, , ,

______ ~ ••• ~ • _~ ~ ••• __ • :- "T,., .... .;

t I I ,

· ,

, ,

, ,

, '

, .

____ •• .L .... _. L •••• ' 1..._ • __ • __ .... •

, I • •

I I • ,

, : h: ' : "h _ ---- -: 100

: L

, ,

___ ... __ ~ L_~ 1... __ ._. ,

. . .

, , ,

, , ,

, . ,

, . ,

· .

· .

· .

, .

, ,

40 I ---r--r' .

30 I I I I I Iii I ! j I I I i I J I I II ii' I I i I I I I r. I I I i .II

o 100 200 300 400 TOTAL PUMPING LIFT (m)

FIGURE 7: Approximate percent pumping submergence for optimum air-lift efficiency (Driscoll, 1986)

then

52 == (L - 93) 100 I L

12

To illustrate an air-lift pumping design, one can look at the following example. A geothermal well is completed to a depth of 500 m with 254 rom (10 in) diameter and 284 m deep casing. This casing has been selected to minimize friction losses and provide adequate clearance for the pump bowls. Static water level is at 26 m. It is estimated that the specific capacity will be about 0.6 Jls/m (3 gpm/ft) for the anticipated pumping rate. It is decided to air -lift pump at a rate of 40 Vs (634 gpm), 20 percent above the design rate of 32 Us (500 gpm). Thus, the water level during pumping will be at about 93 m (305 ft) depth. From this information, it is possible to select the proper size equipment for the air-lift and to estimate the potential efficiency of the system as shown in the following steps:

1.

Determine the diameter of the educator pipe (if required) and the air line. Table 1 shows that a 203 mm (8 in) educator and a 64 mm (3 in) air line should be used for the 254 mm (10 in) casing. If no educator pipe is used, Table 1 recommends the same size of air line to be used.

2.

Determine the lengths of the educator pipe and the air line. When educator pipe is used, the air line will always stay inside that pipe. From Figure 7 for the total pumping lift 93 m (305 ft), the optimum submergence is found to be 52%.

h

%h ;;;; -100

L

(7)

where

or

h : pumping submergence (m)

L : total length of air line (m)

L = h + iJ.H + A

iJ.H : well drawdown (m)

A : static water level (m)

==>

L = 194 m

Which leads to selecting setting depth of educator below the end of air line at about 195 m.

13

3. Determine air volume requirements. For a total pumping lift of 93 m (305 ft), Figure 6 shows that 0.007 m3ts (1 cfm) is required to pump 1 Us (1 gpm). Thus, 0.3 m3ts (634 cfm) are required to pump 40 lis (634 gpm).

4. Determine whether the compressor has sufficient pressure to initiate flow in the air line.

Minimum pressure requirement, P min' (bar)

L-A P.» =-

mID 10.2

(8)

where 1 bar corresponds approximately to 10.2 m of water column. Then

P min = (194 - 26) / 10.2 = 16.5 bar

So an air pressure of at least 16.5 bars (239 psi) will be needed to start the air-lift for the starting submergence selected. As drawdown develops, the pressure requirement drops substantially because the head acting on the air line decreases. In general, air-lift pumping will be most efficient when the static water lever is high, the casing diameter is relatively small, and the pumping depth is not excessive in relation to the pressure capability of the compressor.

During air-lift pumping tests, flow rate, water level or drawdown and temperature are measured. A weir is usually used for measuring the flow rate. Water lever or drawdown in the well can be measured by a pressure gauge at the wellhead connected to a capillary tube, which is installed inside the casing well below the water level or at similar depth as the educator pipe. The water level can be calculated by

H :; L-Ppg

(9)

where

H : water level (m)

L : capillary tube length (m)

P : reading pressure (bar)

p : density of water (kg/nr')

g : acceleration of gravity (m/s2)

224 Injection tests

Geothermal wells are commonly tested at the end of completion by injection tests. The principle is the same as for pumping tests, but the direction of water flow is reversed. The drilling rig pumps are generally used for injection tests. The water is injected through the drill pipes and the pressure (or water level) measured. The flow rate is measured by counting the strokes of the pumps. The water level will increase up to the condition of the reservoir during the test. The data is plotted and analyzed in the same manner as the pumping test in sections 4.2.1 and 4.2.2. The purpose of this operation is to determine the permeability, storativity, laminar and turbulent drawdown for the well.

14

23 Downhole pumps for geothermal wells

Basically, there are two types of deep well pumps most commonly used in geothermal wells, i.e. lineshaft turbine pumps and submersible pumps, the difference being the location of the driver (Rafferty and Culver, 1989). For a lineshaft pump, the driver is usually an electrical motor, mounted above the wellhead and drives the pump through a long shaft. In a submersible pump, the driver (a long, small diameter electric motor) is usually located below the pump itself. The pump is coupled to the driver through a relatively short shaft with a seal section to protect the motor from the well fluid.

Vertical lineshaft turbine pumps in deep well settings have two definite limitations. They must be installed in relatively straight wells and the economical setting limit is not more than 240 m. In geothermal wells, the temperature and chemical content of water are the main problem in selecting the proper material of component parts. The lineshaft material generally used is carbon steel 1045 (AISI C.I045). In some cases the lineshaft must be changed due to the condition of geothermal water. The bowl unit material generally used is cast iron, but special materials can be ordered. The impeller materials generally used are bronze (B 584-838) and cast iron. The most critical part of the pump for geothermal applications is the shaft bearings which are made of teflon (PTFE 1491 N).

r;-r::::! JHD H$p 9000 UK LJ:...J 9'.100673 T

assembly

Bowl

Strainer

FIGURE 8: Basic components - enclosed lineshaft pump (Peabody Floway, 1988)

15

3. DEEP WELL LlNESHAFT PUMPS

3.1 General

Pumping is required to produce water from geothermal wells and to deliver it to the end user. For certain geothermal fluids, such as those with high carbon dioxide content, corrosion and scaling can be reduced by maintaining the water under pressure.

3.2 Selection of a deep well pump

Selection of a pump type (Figure 8) is dictated by the setting depth, well design, well deviation, temperature, type and size of driver, based on lowest life cycle costs, repair costs, and availability of spare parts.

The lineshaft vertical turbine pumps have two basic methods for lubrication. One is the enclosed lineshaftconstruction where oil or filtered water as a lubricant is supplied to the lineshaft bearings through the enclosing tube. This report is written to cover this latter type of construction. The second method is open lineshaft construction which uses the pumped fluid directly for lubrication. The lineshaft and bearings are open to the pumped fluid in this case.

~ JHD HSP 9000 UK L....t:J 91.10.0674 T

A

B

FIGURE 9: A) Semi-open impeller and B) enclosed type impelJer (Peabody Floway, 1988)

Two types of impellers are available "semi-open" and "enclosed" (Figure 9). The arrangement of the bowl assembly is identical with either semi-open or enclosed type requiring only that the intermediate bowls and suction bell be designed to provide sealing surfaces for the type impeller used For geothermal pumps enclosed impellers are used. The pump impellers are connected to the bowl shaft by tapered lock collets or collet and key with locking screw. The shaft and aU rotating parts are supported by the thrust bearings of the hollow shaft motor or separate thrust bearing assembly. There is an impeller adjusting nut at the top of the hollow shaft motor assembly, or a three or four piece coupling with adjusting nut for solid shaft driver arrangements.

16

The column pipes assembly and discharge head are selected for suitable size, length and type for assembly arrangements. The pump size recommended by FLOW A Y vertical turbine pumps (Peabody Floway, 1988) for different flowrates are shown in Table 2 (see also Figure 10).

TABLE 2: Recommended flow ranges for vertical turbine pumps (Peabody Floway, 1988)

Pump size Column size Flow ranges recommended
6~ Column pipe
6" and 8" 2" Enclosing tube . 14·40 1/s
1 3/16" Lineshaft
8" Column pipe
10" 2 1{2" Enclosing tube 50 - 75 lis
1 11/16" Lineshaft
10" Column pipe
12" 2 1{2" Enclosing tube 85 - 1101/s
1 11/16" Lineshaft .. "

~

\

50

4) ,

o

• FLO WAY VERTICAL TURBINE PUMPS 2900 RPM: 6JKH, BJKH, 10LKM 1450 RPM. 12LKH, )2FKL

40 60 ab'

FLOW RATE (l/s)

20

FIGURE 10: Multi pump performance curves

3.21 Borehole

The borehole data that must be known are

C1 : well drawdown coefficient (m/(Vs)2)

t : temperature of water (0C)

Kv : total static water level (m)

3.22 Pump performance curve

The equation of the pump curve (pump pressure versus output) is

This information is specific to each pump and in order to simplify the solution of the equations, the performance curve is approximated with two straight lines (see Figure 11 and Table 3):

(11)

where i = 1 i=2 QB

: ifQ<QB : ifQ~QB

: discharge at point of intersection of the two lines (l/s)

: total dynamic pressure head (m) : number of impellers (stage)

This approximation is fair1y accurate around QB

17

r.r::I JHO HSP 0000 UK L.:t:J 9!.11J_0678 T

5°1

_.-.,

540

'" P...

Q ~

~ 30

~

~

~ 20 Q

.....:I

<r:

E--<

~ 10

Po: 33,4 - 0.3l6 Q

P,,: 41.97 - 0.560 Q

10 20 30 40 50 FLOWRATE Q(ljs)

60

FIGURE 11: &JKH 2900 rpm pump performance curve (Floway, 1970)

TABLE 3: Coefficient of pump performance curves

Coefficient of pump performance curves
PUMP TYPE
Yll Y21 Y12 Y22 QB
FLOWAY 8JKH 3000 rpm 0.316 33.40 0.560 41.97 36.0
FLOWAY 6JKH 3000 rpm 0.240 17.72 0.763 22.89 9.8
FLOWAY 12DKH 1500 rpm 0.139 20.55 0.206 24.78 62.0
FLOWAY 12FKH 1500 rpm 0.066 18.76 0.136 25.83 99.0
FLOWAY lOHKH 1500 rpm 0.040 10.04 0.204 18.36 45.0
FLO WAY lOLKM 3000 rpm 0330 59.60 1.179 99.90 49.0
PEERLESS 8HXB 0.212 27.04 0.528 42.56 50.0
F.MORSE 8HC 3000 rpm 0.286 31.20 0.546 41.55 39.5
FMORSE 6HC 3000 rpm 0.181 13.97 0.613 18.66 11.0 3.23 Pump pressure break down

(12)

where

PH : discharge head pressure (m)

Kv : total static water level (m)

KN : well drawdown (m)

PF : pressure loss in column pipe (m)

Pd : dynamic pressure loss in column pipe (velocity head) (m)

P, : density of water at pumping temperature (kg/m2)

where L

Q

C1 Cz V

g DDI DSU

18

K "" C Q2

N 1

: length of pump setting (m) : well discharge (Us)

: well drawdown coefficient (mJ(Us)2)

: pressure loss in column pipe coefficient (m/(l/s)2/100 m of column) : velocity of water in column pipe (m/s)

: acceleration of gravity (m/s2)

: inside diameter of column pipe (in)

: outside diameter of enclosing tube (in)

(13)

(14)

(15)

Equations 12 to 15 are solved together in order to calculate the only unknown parameter Q:

where

-b±-'b2-4ac Q "" __ V-'-- __

2a

3.24 The minimum submergence of bowl unit

In order to prevent cavitation, the minimum submergence is calculated by:

where hmin : minimum submergence of bowl unit (m)

Pa : vapour pressure (bar)

Pu : atmospheric pressure (bar)

Pr : density of fluid at temperature T °C (kg/m3)

PZQ : density of water at 20°C (kglm3)

NPSHR: net positive suction head required (m)

g : acceleration of gravity (m/s2)

(16)

(17)

19 3.2.5 The minimum length required of column

The minimum length of the pump column in the well is: z._ = Kn + K}12 + KN + hmm.

where KYl KV2 Ky

(18)

: length above sea level to well head (m)

: length below sea level to static water level (m) := Kn + KV2

33 Adjustment of the impellers

A deep well turbine pump is characterized by its long shaft, enclosing tube and column pipe hanging in the well (Figure 12). To ensure trouble free running of this system the net elongation of shaft and column should not exceed the end play (axial clearance) provided for the impellers inside the bowl unit assembly. Following calculations of the impeller movement in the bowl unit assume that the whole pump unit is kept in thermal equilibrium and that the enclosing tube does not share load with the column pipe. The impellers must be adjusted in such a way that during the operation, no rubbing will occur either on the top or bottom. As already selected then, these calculations are for an enclosed line system.

33.1 Elongation of shaft due to hydraulic thrust

The hydraulic thrust on the shaft is calculated by

(19)

: hydraulic thrust on shaft (N) : thrust constant (lb/ft)

: total dynamic pressure head (m)

The pump shaft will elongate because of the hydraulic thrust on it:

TL E :: 9.541·10~-A-

A D2

6

(20)

: elongation of shaft due to hydraulic thrust (mm)

: length of pump shaft (assumed the same length of pump setting) (m) : diameter of pump shaft (in)

332 Elongation of the column due to hydraulic thrust

The total hydraulic thrust on column is equal to the thrust due to total pressure head (PD) on the column system minus hydraulic thrust on the shaft (reaction force):

(21)

or

where

I.T=l JHO HSI> 9000 UK LJ:J 91.10.0677 T

a. oJ

Q ~

II :c

~

II

c a.
c
E ::.::
:l
"8 -=b
r.l
" ::.::
-;;; ""
.s ~
:: r::
"
~
:: Cl
e, c,
II c
E
M "
~ "0
"
"
-0
-~
"
0
::
~
<l
"- d: 20

Deep well pump

Motor

FIGURE 12: Diagram of definitions

11 1

DSU/DSI tl>-

~

DDU/DDI

:: ~

Sea level

Static water level

Pumping water level

Pump

TR : total hydraulic thrust on column pipe (N)

A : effective cross-section area of column pipe (in2)

DDI : inside diameter of column pipe (in)

DSU : outside diameter of enclosing tube (in)

(22)

21

The column pipe will be elongated because of the hydraulic thrust on column (TR):

E == 9.54'10-6 TRL

Rl DDUz-DDI2

where

(23)

ERl : elongation of column (mm)

DDU : outside diameter of column pipe (in)

333 Column shortening due to internal pressure

When the pump is in operation, the column pipe is under internal pressure by the pumped fluid, being greatest at the lowest point. This pressure shortens the column pipe:

where Em L

1.37·1O-S(2PD - LPr '10-3)L

En '" ---------

(DDUIDDlf/' -1

(24)

: Shortening of column pipe (mm) : length of pump setting (m)

Net column elongation is:

where ER

(25)

: net column elongation (mm).

Net relative movement of impellers:

E£ '" EA -ER = EA -ERl +ER2

where EE

(26)

: net relative movement of impellers (mm)

EE must not exceed the bowl unit axial clearance by some safety margin in order to guarantee problem free operation.

The total weight of pump shaft and impellers can be calculated by:

Ts - 41.4D/L+T[

(27)

: weight of pump shaft (N) : weight of impeUers (N)

Total thrust load on motor: Total thrust on shaft, that is carried by the motor thrust bearing, is:

(28)

: total thrust on shaft (N)

: hydraulic thrust on pump shaft (N)

: weight of pump shaft and impellers (N)

22

33.4 Required size of the motor

The high thrust vertical induction motors of the hollow-shaft type are usually used to drive the deep well pumps (in some cases solid shaft motors are used). The power required size of the motor can be calculated by

where AM AD AL AA PI

Q

TID

n

f

(29)

(30)

(31)

A - fL·1Q-2 A -

(32)

: power required by motor (BHP) : power required of pump (BHP)

: power loss in motor thrust bearings (HP) : power loss of pump shaft (HP)

: density of water at T ClC (kg/m3) : well discharge (Us)

: efficiency of pump (%) : rotation of pump (rpm)

: coefficient of friction loss in lineshaft bearing (HPIL = 100 m)

The power consumption of motor is defined by the equation below AM

A ;:-

N '11M

(33)

and kilowatt input to motor by

where

(34)

: power consumption of motor (HP) : efficiency of motor (%)

: voltage per leg applied to motor

: amperes per leg applied to motor : power factor of motor

: factor for 3-phase motor (this reduces to 1 for single phase motors)

3.4 Material selection

In this section the subject is material selection for the deep well pump unit. The main components and their sizes are highlighted.

23

Pipes

Specification of material: Seamless black steel tube according to DIN 244811629/page 3, mat. st 35, mill test certificate to DIN 50.049/2.2, plain ends, in exact lengths of 3100 mm ± 5 mm tolerance (Table 4).

TABLE 4: Sizes of column, coupling and enclosing tube

COLUMN PIPES
Nominal dia. (in) Outside dia. (mm) Thickness (mm)
5 139.7 6.3
6 168.3 7.1
8 219.1 8.0
10 273.0 inn
COUPLING
5 159.0 12.5
6 193.7 17.5
8 244.5 17.5
10 298.5 17.5
ENCLOSING TUBE
2 60.3 63
21;2 76.1 8.0 Capillary tubes

Specification of material: Welded black steel tubes according to DIN 2441, st 33 DIN 1626, with threads and muffs. Pipes outside protected against rust in fixed lengths of 3100 mm ± 5 mm, bundled. Sizes 1/4" or 3/8" black.

Teflon bearings for Iineshaft

Specification of material: Teflon type PTFE 1491 N. See Table 5.

TABLE 5: Size of bearing stock material required for fabrication

SIZE OF SHAFT (in) OD/ID (mm) LENGTH (mm)
1 3/16 shaft with sleeve 45/35 130
1 3/16 shaft without sleeve 45/28 200
1 11/16 shaft with sleeve 58/47 170
1 11/16 shaft witbout sleeve 60/40 200 Sleeves of shaft

Specification of stock material: Seamless steel tubing of AISI 304 stainless steel, 4-5 m.

SIZE OF SHAFT (in) SIZE OF SLEEVES OD/ID (mm)
1 3/16 40/25
1 11/16 56/40 Lineshaft pump

Specification of material: Shaft materials of deep well pumps are in lengths 3100 mm, bright precision, cold drawn, stress relieved ground and polished bars of steel C 45 (AISI CI045). Maximum diameter tolerances + 0-0.06 mm (h 9 according to DIN 7157). Shaft straightness must be within 0.08 mm total indicator reading and end sawn straight.

24

Shaft size stock material: 1 3/16 in (30.16 mm), 1 11/16 in (42.86 mm), 35 rom, 45 mm (Tables 6 and 7, Figures 13 and 14).

TABLE 6: Calculated weight of components

SIZE THICKNESS WEIGHT LENGIH WEIGHT
(mm) (kg/m) (m/unit) (kg/unit)
Column 10" (273.0 mm) 10.0 64.9 3.0226 196.11
Column 8" (2I9.1 mm) 8.0 41.5 3.0226 125.44
Column 6" (168.3 (mm) 7.1 28.3 3.0226 8554
Enclosing tube 2 1/2" 7.1 12.1 15208 18.40
Enclosing tube 2" 63 8.42 1.5208 1281
Enclosing tube 2" 5.6 7.53 1.5208 11.45
Sbaft 1 3/16" (2 '* 0.31) + (5.617 '* 3.042) 17.71
Shaft 1 11/16" (2 '* 0.55) + (11.34 * 3.042) I 35.60 TABLE 7: Total weight of components in kg/unit (3.042 m)

WEIGHT OF COMPONENT
ITEMS 10" 8" 6"
21a" 21{2" 2"
1 11/16" 1 11/16" 1 3/16"
Shaft 35.60 35.60 17.71
Shaft coupling 0.79 0.79 0.45
Enclosing tube x 2 36.80 36.80 25.62
Bearing x 2 2.09 2.09 1.32
Centralizer 3.01 1.30 0.98
Column 1%.17 125.44 85.54
Column couolin_z 15.0 135 6.1
Total weight 289.5 215.52 137.7 r.-r:::I JHC HSP'9rJ<XlUK L.:.....tJ 9IJO.OO18T

r.r::l JHO I-fSP 9000 UK ~ 91.1Q..06?9T

4000

100

20

t

<t

::c: 2000 en

c...

o

E-

~ 1000

W

~

M

e:: 80 o,

Z

::E

;:>

...J

o 60 U

~

o

~ 40 o:l

::E

;:>

Z

O~~~~~~~~,~~~~~~~

o 50 100 150 200 250 300

LENGTH OF SHAfT (m)

FIGURE 13: Weight of shaft with length

- 6" K Z" x 1 3/16"

<HH><H> 8" x 2 1/2" x 1 11/16" ~ 10" < 2 1/2" x 1 II/IS"

Remar-k; Not. including bowl unit Of,~~~~~~~~~~~~~~I~

o 5 10 15 20 25 30 TOTAL WEIGHT OF STRING (ton)

FIGURE 14: Total weight of deep well string

3.5 Maintenance

25

A periodic inspection is recommended as the best means of preventing breakdown and keeping maintenance cost to a minimum. A change in noise level, amplitude of vibration, or performance can be an indication of impending trouble. Variances from initial performance will indicate changing system conditions or wear or impending breakdown of the unit.

A periodic once-a-week inspection is suggested for all units. During this inspection, the pump and driver should be checked for performance and change in noise or vibration level, loose bolts or piping, dirt and corrosion.

Impeller re-adjustment after thermal equilibrium is reached if properly set at initial installation of pump. After extended operation under abrasive conditions, the sealing faces will wear, causing a reduction in performance.

Pump lubrication is very critical and requires careful monitoring. Adjustment of packing gland is necessary to prevent excessive leakage. The driver lubrication will require periodic attention. Refer to the driver instruction manual for recommendations.

3.6 Repairs

Before puUing the bowl unit from the well, the installation and operation manual should be reviewed for causes of reduced performance not related to the condition of the bowl assembly. Some of the possible causes of reduced performance could be:

- Incorrect lateral adjustment. (This is especially critical with units utilizing semi-open impellers)

- Well conditions, such as gas or air in the water.

- Change in operating conditions such as increased pumping fluid level or increased

discharge head requirements.

- Slow motor speed due to overload, low voltage or low frequency.

- Incorrect direction of rotation.

- Strainer clogged or suction "sanded in".

~ Inadequate submergence of pump suction.

Disassembly of the unit is basically the reverse of the assembly procedure. Handle all parts with care to prevent bending of shafts and damage to machined surfaces.

Check shaft for straightness; shaft must be straight within O.OOS (0.127 mm) total indicator reading. Figure 15 indicates the recommended method for checking shaft straightness. If the shaft is not straight it must be straightened or replaced. If the deflection is gradual over the considerable length, the shaft can usually be straightened by supporting the crooked section on two blocks straddling and applying pressure to the high side to deflect the shaft to the opposite direction. If the shaft has a sharp crook (dog-leg), it is recommended that the

r.n JHD HSI> 9000 UK ~ 91.10.0680T

Itial m, dicator

. .#Sh~

Shaft roll

"Aft

FIGURE 15: Checking shaft for straightness (Peabody Floway)

26

shaft be replaced since the shaft will not always remain straight even if satisfactorily straightened

After disassembly, all components should be thoroughly cleaned and examined for physical defects, wear, corrosion and damage. The bowl assembly becomes worn when operated a long time. The nominal and maximum allowable of parts are shown in Table 8.

Parts showing signs of damage, cracks or excessive wear should be replaced. When repairing a pump that has been in service for several years, the physical condition or straightness of all parts such as cap screws, bowls, threads, etc., must be carefully checked to be sure these parts can continue to perform their function without failure.

TABLE 8: Deep well pump material nominal size and maximum allowable wear

TYPE MATERIAL NOMINAL (mm) MAX. ALLOWABLE WEAR
(mm)
Shaft sleeve OD 48.00 ·
Bearing ID 48.35 48.78
Shaft 1 11/16" Shaft OD 42.86 ·
Tube adaptor bearing ID 43.00 43.43
Stuffina box bearing lD 43.10 43.53
Shaft sleeve OD 35.00 -
Shaft 1 3/16" Shaft OD 30.16 -
Tube adaptor bearing ID 30.30 30.61
Stuffina box beartnz ID 30.40 30.71
6JKH 1"
Shaft OD 24.50 ·
Imoelter skirt OD . -0.31
8JKH 1 3/16"
Shaft OD 30.16 ·
Bowl units from
Floway Imoeller skirt OD - ·0.31
12DKH 1 11/16"
i Shaft OD 42.86
-
Impeller skirt OD - ·0.33
12FKH 1 11/16"
Shaft OD 42.86 ·
Impeller skirt on . -0.41 3.7 Engineering information

This section will give information about limitations of mechanical and electrical equipment for deep well pump assemblies (Tables 9, 10, 11 and 12).

Standard bowl assembles are generally designed for non-corrosive, non-abrasive service over a

27

wide temperature range (Table 9). Published performance CUIVes are based on a maximum fluid viscosity of 50 SSU. The maximum pump speed is also limited by lineshaft RPM limitations, critical speed, and/or submergence (NPSH) available to prevent cavitation.

TABLE 9: Limitations of standard bowl assemblies (Peabody Floway, 1988)

SIZE SHAFT MAX. STO. MAX. AXIAL THRUST IMPEllER EYE SPECIFIC WR~0 MAX.
& SIZE MAX. PRESSURE AXIAl..@ CLEARANCE FACTOR WEIGHT AREA SPEED (WET@ SPHERE
TYPE (IN.j(i) STGS.® P.8.J.@ CLEARANCE (bymach·l@® (K)@ (L8S.) (IN2"j (Ns) 1.08.G.) {IN.}®
4HOLl 'I. 32 500 ... ... 1.6 0.9 1.4 2526 0.008 '10
4HOl 0/. 32 500 .'. 2.4 0.9 2.0 2750 0.009 '/'6
4HOH 3J. 32 500 ... ... 2.2 0.9 2.4 3139 0.008 5;'6
6lKM 1 32 606 '/. 7/16 2.1 2.4 1.3 2155 0.046 '/'6
6JOLl 1 33 489 3.6 T.9 2.7 2355 0.047 '/.
6JOL 1 33 489 ... 3.6 1.9 3.9 2350 0.051 %
5JOH 1 33 489 ... ., . 3.6 1.9 4.9 2710 0.046 'h
6JKL 1 33 489 0/. ';' 3.6 2.9 6.5 2680 0.059 'h
6JKM 1 33 489 % '12 3.6 2.9 6.5 2800 0.057 %
6JKH 1 33 489 % '12 3.6 3.0 6.5 2740 0.055 'h
8LKL 1";"6 35 429 'h % 4.2 4.5 4.1 2510 0.132 %
8LKM P/,e 35 429 "/'s "/'6 4.2 4.5 4.8 2370 0.132 %
saoi. P/'6 29 414 ... ... 4.4 3.8 7.65 2690 0.108 'I.
8JOH 19;'6 29 414 ... ." 5.5 3.4 7.65 2550 0.102 1'.li6
SJKL 1';'6 29 414 "/'6 '/. 38 5.2 11.6 2580 0.210 'I.
SJKM 1'A6 29 414 9;"6 '/. 4.7 5.4 11.8 2660 0.210 3J.
8JKH PAs 29 414 91.", % 4.7 5.5 11.6 2700 0.215 ";'6
8FKH 1:Y,6 31 310 ',4 'h 5.5 5.7 11.2 4030 0.178 '12
8XKl 1'A8 19 575/1000 7/l6 ... 2.4 4.3 2.9 1500 0.129 %
8XKH 1:Y'8 19 575/1000 ,/'S ... 2.4 4.2 2.9 1510 0.126 %
8XKN PI,. ... '" 7/'6 ... 7.2 6.1 10.5 1853 0.133 %
10LKL 1"" 19 488 ';' 'Ie 5.2 9.0 10.1 1690 0.675 '12
10LKM 1'12 19 488 Th % 5.2 9.0 10.1 1715 0.692 '12
1000l 1'12 24 383 ... ... 10.4 6.0 18.2 3140 0.381 1
1000M 1'12 24 383 ... 10.4 6.0 18.2 2980 0.381 7;;'
1000H l'h 24 383 ... ... 10.4 6.,2 18.2 2970 0.376 7,."
10DKl 1% 24 383 9/16 3J. 8.3 10.8 18.,2 3150 0.598 1
10DKM 1% 24 383 9;', 3/. 8.3 10.8 18.2 3110 0.605 %
100KH 1% 24 383 9/,. 3/. 8.3 10.9 16.2 3160 0.613 '18
10BKL I';, 29 410 7As % 10.0 10.75 20.9 3850 0.675 1
10BKM 1% 29 410 7/tS % 10.0 9.75 20.9 4260 0.609 1
10BKH 1% 29 410 7A6 % 10.0 10.75 20.9 4260 0.675 1
10FKH "1,{,6 24 333 9/'6 % 13.3 9.0 26.B 5800 0.711 1
10XKL 1'12 14 500/1000 'I. . .. 4.6 8.1 7.9 1630 0.409 '12
10XKH 1% 14 50011000 'l2 ... 4.6 8.7 7.9 1560 0.439 %
10XKN 1% 1 NA 'I> ... S.8 8.8 17.3 1650 0.379 'I,.
11XKL r1'tA6 10 75011500 '/. ... 5.7 11.5 9.5 1595 0.709 '12
l1XKH 111;;6 ... 750/1500 0/. . .. 5.7 11.7 9.5 1545 0.721 %
l1XKN P1fi6 1 NA '/. ... 10.3 10.6 20.9 2025 0.580 '12
12LKl 1 ";'. 28 405 % Til! 8.0 19.9 15.0 - 2680 1.788 'j.
12LKM 1""/'115 28 405 '12 % 8.0 16.9 17.1 2280 1.594 'Ie
12LKH 1 "/,. 28 405 '/2 '/. 6.0 17.1 17.1 2370 1.537 %
12DOL 1"/16 25 352 ... ... 15.0 11.0 26.4 2900 0.711 1 'I.
12DOM 1''';'6 25 352 ... 15.0 11.5 26.4 2860 0.734 1%
12DOH 1'1Ju:1 25 352 ... 15.0 12.0 26.4 2845 0.739 1
12DKL 1 "/'6 25 352 % '>;i. 12.0 16.5 26.4 3180 1.015 1 'I.
12DKM 1 ";'. 25 352 % 15116 12.0 15.0 26.4 3110 0.993 1'/0
12DKH .,-n/U!i 25 352 % 'I~/,ofi 12.0 15.0 26.4 3050 0.997 1
12FKL 111!Ul 25 267 'h 1 15.5 23.6 32.2 3370 2.164 1
12FKH 1"/1115 25 267 'h 1 13.5 18.6 32.2 3770 2.178 1
13XKL 115/16 6 75011500 % 8.1 19.3 13.7 1590 1.687 %
13XKH 1"/,. S 75011500 7/. . .. 8.1 19.8 13.7 1540 1.730 %
13XKN 1 'Sf,. 1. NA 7,4 16.6 23.0 33.7 1610 2.059 % 28

The horsepower loss due to mechanical friction of shaft rotation in the column (AA in Equation 32) may be determined from Table 10 below. For an enclosed lineshaft with a flooded tube or the enclosed lineshaft using water injection lubrication, use two times the factor values listed in the table.

TABLE 10: Mechanical friction in BHP per 100 m of lineshafts (Peabody Floway, 1988)

SHAFf RPM OF SHAFT
SIZE (in) 3460 2900 1760 1450 1160 960 860 720 690
3/4 0.60 0.51 0.31 0.26 0.20 0.17
1 1.05 0.87 0.53 0.44 0.35 0.29 0.26 0.25
1 1/4 1.60 133 0.79 0.64 0.52 0.44 0.39 0.34
1 1/2 2.20 1.90 1.14 0.96 0.74 0.63 0.56 0.47 0.44
1 11/16 2.81 2.36 1.43 1.18 0.94 0.78 0.70 059 0.56
1 15/16 1.83 1.51 1.21 1.00 0.89 0.75 0.72
21/4 2.40 200 1.55 1.35 1.20 LOO 0.96
27/16 2.90 2.39 1.91 158 1.42 1.19 1.14
2 11/16 334 2.75 2.20 1.82 1.63 138 1.31
215/16 3.50 2.77 2.28 2.00 1.74 1.66
3 3/16 4.04 3.28 2.72 2.46 2.08 1.92
37/16 3.75 3.08 2.83 2.45 2.30
3 11/16 4.10 3.46 3.00 2.62 2.47
3 15/16 4.80 3.89 3.48 2.94 2.77 The shaft horsepower ratings are shown in Table 11.

TABLE 11: AISI C-1045 carbon steel shaft horsepower ratings (Peabody Floway, 1988)

SHAFT DlAM:ETER (inches)
RPM 1 15{l6 211M 311/H
3f4 1 13/16 I1f4 11/2 1 11/16 2114 27/16 2 15/16 33f16 37{16 315/16
3500 55 137 222 264 475 706 1094 168
2900 46 113 184 219 393 585 906 1394
1770 28 69 112 134 240 357 553 851 932 1280
1450 23 57 92 109 197 293 454 697 764 1048 1389
1180 19 46 75 89 160 238 369 567 622 853 1131 1467 1859 2315 2843
980 15 38 62 74 134 198 308 471 516 709 939 1218 1543 1923 2361
880 14 34 56 66 120 178 276 423 464 636 843 1093 1386 1727 2120
730 29 46 56 99 146 228 351 385 528 699 907 1150 1432 1758
705 28 45 54 96 142 220 340 371 510 676 875 1110 1383 1699
590 38 45 80 118 184 284 311 4Z7 565 732 929 1158 1421
505 33 39 68 102 158 243 266 365 484 627 795 990 1217
Max.thrust (100) 3000 4000 5000 5000 10000 10000 15000 20000 20000 20000 25000 25000 30000 35000 40000 Column friction loss is loss due to fluid flowing between the column surface and the enclosing tube. This column friction loss is used for calculation in Equation 14 as shown in Table 12.

29

TABLE 12: Friction loss (m) per 100 m of column (Peabody Floway, 1988) (Open or enclosed lineshaft design)

COLUMN ANO SHAFT SIZE (Inchesl
2% 3 4 .5 6 B 10 12
'I. :v. :v. 1 1 'I. '/. , 1 'I .. '/, I ,';'" l'k I 1 'I. l'h 1lsj..e 1 'I. l'h 1IS/t" 2'1. 11/. 1'1.> 1·V'6 ;2V" 27/16
GPM 111/'5 1"/i8 ,1'1'1[1 2'\',\1:
10 1.2
15 2.0
20 2.8 1.0
25 3.5 1.4
30 4.2 1.9
40 5.' 3.' .6 1.2
50 6.6 4.5 .7 .9 1.7
60 9.0 6.1 .9 1.2 2.3
70 6.5 1.2 1.6 2.9
80 9.9 1.5 1.9 3.7
90 12.0 1.8 2.4 4_'
100 14.0 2.2 2.8 5.3 .6 .7 1.0
'25 3.2 4.2 7.8 .B 9 1.4
,SO 4.4 5.8 1.2 1.3 1.9
175 5.8 7.5 1.6 L7 2.5
200 7.3 2 .. 0 2_2 3.1 .8 .7 1.0 ,.4
225 2.9 2.8 3.9 1.0 .9 '.2 1.7
250 3.1 3.3 4.7 " 1.1 1.' 2.0
275 3.7 3.9 5.6 1.3 1.3 1.7 2.4
300 4.3 4.5 6.4 '.6 '.5 2.0 2 .. 8
325 5.0 5.4 7.4 1,6 1.7 2.3 3.,2
350 5.B 62 8.4 2.0 2.0 2.6 3.6
375 6.6 7.0 2.3 2.2 2.9 4.1
400 7.4 7.9 2.9 2.5 3.3 4.5 .6 .7 1.0
450 9.3 3.5 3.' 4.1 5.7 .8 .9 1.3
SOO 4.2 3.7 5.0 6.9 '.0 ,_, 1.5
550 4.9 4.4 5.8 1.1 1.2 1.3 1.8
600 5.7 5.2 6.B 1.3 I.' '.5 2.1
650 6.6 6.0 1.5 1.6 1.8 2.5
700 7.5 1.7 1.9 2.0 2.a
750 2.0 2.1 2.3 3.Z
800 2.2 2.' 2.6 3.6 .6 .7 .8 1.0
850 2.5 2.7 2 .. 9 4.0 .7 .8 .9 1.1
9QO 2.8 3.0 3.2 4.5 .8 .8 1.0 1.2
950 3.1 3.3 3.6 4.9 .9 .9 1.1 1.3
1000 3.4 3.6 3.9 5.4 1.0 1.0 1.2 1.4 A .4 .4 .5
lZ00 4.8 5.' 5.6 7.6 1.3 1.4 1.7 2.0 .6 .6 .6 _7
1400 6.4 6.8 7.4 10.0 1.8 1.9 2.2 2.7 .7 .8 .8 1.0 1.2
1600 8.3 B.B ·9.5 2.3 2 .• Z.8 3.4 .9 1.0 1.1 1.2 1.4
1800 10.3 11.0 11.9 2.9 30 3.5 4.3 1.2 ;.,2 1.3 1.5 1.8
2000 126 3.5 3.7 •. 3 5.2 1.4 1.5 1.6 I.S 2.1
2200 15.1 4.2 4.4 5.1 6.1 1.7 1.8 1.9 2.1 2.5
2400 49 52 6.0 7.2 2.0 2.1 2.3 2.5 3.0
2600 5.7 6.1 7.0 B.2 2.3 2.5 2.6 2.9 3.5
2800 6.6 7.0 B.O 9.6 2.7 2.8 3.0 3.3 '.0
3000 7.5 7.9 9.' 10.0 3.' 3.2 3.4 3.S 4.5
3200 3.5 3.6 3.9 4.3 5.1
3400 3.9 4.1 4.3 '.8 $.7
3600 '.3 4.5 4.8 5.3 6.4
3800 4.8 50 5.3 5.9 7.1
4000 $.3 5.5 5.9 6,4 7.8
4200 5.8 60 6.' 7.1 8.8
4400 6.3 6.6 7.0 7.7 9.9
4600 6.8 7.2 7.6 8 .•
4600 7.' 7.8 8.3 s.o 30

4. PUMP DESIGN AND SELECTION

UTlllZATION OFTIIE WELL TEST·1 wrm: THE lEMPERATURE 13O"C KNOWN DATA:

1.

Well data.

Depth

Casing diameter Casing length

Elevation (KVl) (m.a.s.l.) Water temperature

Density of water at zoce (Pzo) Density of water at 130°C (Pt) Wellhead pressure (PH) Vapour pressure CPa)

Minimum atmospheric pressure (Pu) Acceleration of gravity (g)

: 928m :230m : 284 m :26m

: 130°C

: 998 kg/m3 : 935 kg/m3

: 44 m (4.4 bar) : 2.7 bar

: 0.96 bar : 9.81 m/s2

Drawdown characteristics of the well were measured by air-lift pumping. The temperature of the water during the test was 66°C and the data from well testing is as follows:

Flow rate 15.01/s 27.5 1/s 40.1 lis

Drawdown :9.3 m

: 20.1 m

: 51.1 m

2. Geothermal reservoir data. The water table in the reservoir is estimated to vary between 60-100 m b.s.1. (KV2) during the operation period.

3. The need for the water is approximately 40 1/s for our process.

CALCULAlE TIlE FOLLOWING AND DElERMINE:

I - Well drawdown coefficient (el)

n - Suitable pump to approximately pump at 40 l/s. Draw the curve for Q as a function of Kvz- The variation as a function of time influences the selection of the pump by seasonal variation and gradually decreasing water table.

III - Adjustment of the impellers after thermal equilibrium is reached.

IV - Suitable motor size for the pump.

V - Verify that the equipment selected has not reached its limitation.

VI - Minimum lifting capacity of the mobile crane.

Figures 16-17 show the calculation sequence used in this report. The equation number used in each step is shown in brackets (see Section 3).

I.T=1 ·JHD HSI> 9000 UK L....t:J 91.10.0554 H

WELL PUMP TEST (Drawdown)
T, Q, K\( ,Kv
·1 2

CALC. WELL LOSS COEF.
C (2)
I
WELL DATA
.....
~
CALC. ROUGH LENGTH OF
COLUMN AT DIFF. FLOWRATES
Lmin (2.,17,18)

ESTIMATE MAK PROD. RATE
(L min <240m)

SELECT PRODUCTION
CAPACITY Q

PUMP DATA (Catalog)
-
i.>
SELECT PUMP TYPE
(Catalog, Fig.)
I
SELECT COLUMN AND
SHAFT DIA (Catalog, Table)

ESTIMATED TOTAL DYNAMIC
PRESSURE HEAD Po
I (12, 13, 14, 15)
r
'CALC. NUMBER OF IMPELLER
STAGE REQUIRED I FIGURE 16: Preliminary pump calculation

31 SOLUTION:

I - Well drawdown coefficient (el)

Air-lift pumping data is plotted. Flow rate versus drawdown and best fit curve are calculated. The data was provided by the built-in function of the program "GRAPHER" (Golden Software) as shown in Figure 1. Equation 2 for the drawdown

curve is .

where

KN = i1H = well drawdown (m)

Cz = well drawdown coefficient (m/(l/s)2)

giving

KN = 0.0337 Q2

CZ = 0.0337 rn/(I/sl

Rough calculation of the maximum pumping rate by using deep well pump:

From Equations 11 and 18, where we assume the lowest water level in the reservoir (KV2=loo m):

At pumping rate 40 lis:

Elevation (Kvz)

Water table (Kn) maximum depth Well drawdown (KN)

Minimum submergence (hmin) Total depth required (Lmin) Selected length setting bowl unit (add 5% for safety)

26m 100m 54 m 28m

208 m

219m

At pumping rate 60 Vs:

Elevation (Kvz)


Minimum submergence (hmin) Total depth required (Lmm) Selected length setting bowl unit (add 5% for safety)

26m 100m 122 m 28m

276m

289m

At pumping rate 70 Vs:

Elevation (KVJ)


Minimum submergence (hmin) Total depth required (Lmin) Selected length setting bowl unit (add 5% for safety)

26m 100m 165 m 28m

319m

335 m

32

Ul Z o

~

5::J

~::l

::J~ r---------~-------------------

au

r.n JMO HS~ 9OCo UK ~.g1-10.05s:l ...

(/) z o

~

::J

s

-c o

ti:

~ I (/)

o z ~

Z ::!: ::l ..J

o U

INPUT DATA FROM PRELIMINARY PUMP CALC,

Rough calculation of pump length:

The water table in the reservoir is variable over the operating time between 60-100 m, There are two types of variations, seasonal variations and gradual decrease (Figures 18 and 19). These can influence the selection of pump and number of stages. As we are going to pump 40 Vs, the minimum pump length is 208 m. For safety, the length of bowl unit is selected as 219 m (72 column pipes).

CALC, ACTUAL Ft.aw RATE INTERSECTlDN OF PUMP ANO SYSTEM OURVES 0. (' 6)

CA~C., TOTAL DYNAMIC P,S5 OF PUMP Po [11)

CALC. R.EQU I RED COUUMN LEN~ (L > Lm;n) [18)

, CHANGE TO:

1) HYDAUUC BAlANCED IMPElLERS

2) INCREASE SKAFT DIA,

3) INCREASE .AXlAL CLEAAiu'lCE (Deep se "'n9 typo)

CALC, T><RUST B~RI NG LOSS AL (31)

CALC. LI N E S HAFT LOSS A;' (3<)

CALC, HYDAUUC B!lA~ HORSEPOWER! ,0.,0 (30)

FIGURE 17: Detailed pump calculations

n - Selection of a suitable pump

Pumping rate 40 Vs and setting bowl unit at depth 219 m have been selected within the maximum practica11ength of the deep well pumps. From Peabody Floway multi pump performance curves (see Figure 10) we select 8JKH pump type for this application.

33

50 rn JHO HS~ SOOO UK 50
91, J().0S8l T
60 60
----
S E-
N 70 ~ 70
"
::.::: ::.:::
r.-J r.-J
.....:l o--l
c:o 80 ro 80
<>: <>:
s- E-<
n:: 0::
I'i1 90 e:a 90
E-< E-<
<r; <>:
~ ;>=
100 100
SEASONAL VARIATlON GRADUAL VARIATION

11 0 , 11 0 -'h---rrrT"T""""-rr-t--rr-rr-rT"TT"T"T"T-,-,-,r-rrrTTTTTT"T""1""""'--"'"

0,0 0.5 1.0 1.5 2,0 2.5 3,0 3,5 4,0 0.0 0.5 1.0 1.5 2.0 2,5 30 3.5 4.0

TIME (year) TIME (year)

FIGURE 18: Seasonal variation of water table FIGURE 19: Gradual variation of water table

Pump data:

From Peabody Floway vertical turbine pumps, turbine data handbook (2nd ed, 1988)

8JKH, 2900 rpm Thrust constant (K) Impeller weight WR2

Shaft diameter (Do) Maximum sphere Eye area

Maximum head (std. const.)

Maximum number of stages (std. const.) Axial clearance (std.)

Axial clearance (max.)

Bowl diameter

NPSHR (at approximate 30-40 lis)

4.7 lb/ft 5.5 lb 0.215 Ib/ft2 1 3/16 in 11/16 in 11.8 in2

957 ft (292 m) 29 stg

9/16 in (14.29 mm) 3/4 in (19.00 nun) 73/4 in

8 m

A pump performance curve of 8JKH 2900 rpm with two straight lines is used to approximate the characteristic of the curve (see Figure 11 and Table 3).

Selection of column and shaft from Section 3.4:

Column x enclosing tube x lineshaft 6" x 2" x 1 3/16" (DDU = 6.626", DDI = 6.067", DSU = 2.374", DSI = 1.878")

Rough calculation of total dynamic pressure head required:

In order to find out how many impellers (stages) we need for the 8JKH pump, one has to estimate the total dynamic pressure head required to pump 40 lis from the well.

34

(12)

PH = 44 m

Kv = KVJ + KV2 K0 = 26 + 60 KvlJ = 26 + 80 KvC = 26 + 100

= 86m = 106m = 126m

where

A B C

: in case water table KV2 60 m b.sJ. : in case water table KV2 80 m b.s.1.

: in case water table KV2 100 m b.s.l,

NOTE: Each step in the following calculation is made for the above three cases with different water tables in the reservoir. These cases are designated, for example for Pn, as P~ etc.

K "" C Q2

N 1

(13)

= 54.24 m

p - C Q2L/l00

F - 2

(14)

From Table 12, 6" column friction loss, 1 3/16" shaft diameter at 40 1/s (Section 3.7):

PF = 7.5 m / 100 m of column

or for column length 219 m

PF = 7.5 * 219/100

C2 = 16.4 I [402 * (219 / 100)]

= 16.4 m

then

PD = O.229Q2f(DDlz -Dsu2j2 PD = 0.229 '" 402/ (6.067)2 - (2.374)2)2

Pot! == 44 + (86 + 54.24)935 * 10.3 + 16.4 + 0.377 PIfi = 44 + (106 + 54.24)935 * 10.3 + 16.4 + 0.377 PDC = 44 + (126 + 54.24)935 * 10.3 + 16.4 + 0.377

(15)

= 0.377 m

= 191.90 m = 210.60 m = 22930 m

Calculation of the required number of impellers:

From 81KH performance curve at flow rate 401/s, the total dynamic head is 19.6 m per impeller stage.

Number of impellers (Z):

A = 191.90/19.6 B = 210.60/19.6 C = 229.30/19.6

= 9.79 ~ 10 stages = 10.74:::::: 11 stages = 11.70::::: 12 stages

35

Detailed calculations:

We have already selected the 8JKH pump and its length to be 219 m. The selection of impellers required for the pump depends on how the water table in the reservoir (Kv) varies with time. If it decreases gradually, over for example a ten year period, then it is economical, especially if the price of electricity is high, to start with 10 impellers. When the water table nears 80 m.b.s.l. (case B) then one more impeller is added to the pump etc. The size of the motor is, instead, based on the maximum number of impellers to be used. If the water table has a seasonal variation of 60·100 m.b.s.l, and our process needs 40 Us all the time, then we must select 12 impellers for the pump so it can pump 40 Us at the lowest level (case C), The following detailed calculation are shown for case A where KV2 = 60 m. The same calculations were made for case B where Kn = 80 m and C where Kn = 100 m. The results of these calculations are shown in Table 13.

Calculation of actual flow rate:

For the selected pump the actual flowrate is calculated by Equation 16

-b±'b2 -4ac Q = __ v~ __

2a

(16)

where the coefficients a, band care

a = C P '10-3 + C '(L/lOO) + 0.229

1 1 1 (DDI2 _ Dsu2i

Hence,

a ~ 0.0337 * 935 * 10,3 + 0,0047 * (219/100) + 0.229 I ((6.067)2 - (2.374)2)2 = 0.042

The equation of straight line from performance curve where Q > QB is PD == YU-YliQ

Hence,

PD = 41.97 - 0.56Q

where

Yz; = 41.97 Yli = 0.56

This gives

b = 0.56 * 12

= 6.72

and

cA = 44 + (86 '" 935 '" 10,3) - (41.97 * 12)

= . 379.23

Hence,

QA = 442 Us

Calculation of total dynamic pressure head of pump (Equation 11):

PD "" (Y2i-YUQ)Z (11)

p~ = (41.97 - 0.56 * 44.2)12 =

206.6m

A graph illustrating the flow rate change due to ~ 70 water table variation in the reservoir is shown ::.::

c.J

in Figure 20. iI5 80

« E-

o::

~ 90 «

Ol:

Calculation of minimum submergence of pump:

The minimum submergence is calculated with the help of Equation 17.

h. = (P .. -P,)llfl + NPSHR (17)

mm ptg (Pt/P20)

h1flin = (2.7-0.96) lOS I (935 * 9.81)

+ 8 I (935/998) = 2750 m

36

I.T""=l JHD HSP sooo UK L.:....CJ 91.10.0083 T

50

60

E

100

1 1 0 -+r-riTTT1rrTTT~TnTrrrTrrn-rn-nTTTTTTTTTTrT"TT1"T"T"TTnrTTnrITl

39 40 41 42 43 44

FLOW RATE Q O/s)

45

FIGURE 20: Plot showing flow rate versus water table

Calculation of minimum column length:

For setting bowl unit in the well is given by Equation 18.

(18)

L",m4 = 86 + 0.0337 * (44.2)2 + 27.5 =

1793m

Determine the actual length of column pipes:

LminC (Table 13) shows the maximum depth of the three cases. If we select 209 m and add 5 % column pipe then the length of column pipes will be 219 m. The column unit is 3.042 m/unit. To calculate the number of column pipes, 219 m are divided by 3.042 m, which gives 72 units.

The actual column pipe's length is 219 m for setting bowl unit.

ill - Decide the adjustment of the impellers after thermal equilibrium is reached

Calculate the thrust on shaft, column, elongation of shaft and column. Then we will get the relative movement of impeller. If we use a standard 8JKH pump with 19 mm axial clearance and a thrust constant K = 4.7 Ib/f, we find from Equation 26 that the net elongation of the impellers are between 19-22 mm (case A-C). Therefore, we cannot use the standard pump in our application.

The problem can be solved by:

• Increasing shaft diameter (D Q) to decrease elongation and relative movement of impeller.

- Decrease thrust constant (K) by using hydraulic thrust balanced impellers.

- Use a special deep well setting bowl unit with greater lateral clearance. The bowl unit

cost is higher than normal bowl unit.

37

In OUf case, we select the alternative using balanced impellers, 6 out of 12. This means 6 hydraulic thrust balanced impellers and 6 normal impellers. The new thrust constant (K) is 4.7 * 6/12 = 235 Ib/f.

Calculation of hydraulic thrust on shaft (Equation 19):

TA, "" 14.6KPD

T,A"f = 14.6 * 2.35 * 206.6

Calculation of shaft elongation (Equation 20):

E_,.tA = 9.541 * 10-6 * 7,137.0 * 219/ (1.187)2

Calculation of total hydraulic thrust on column (Equation 22):

Tk "" 4.97(DDl'l-DSU'l)PD-TA

T Jt! = 4.97 «6.067f • (2.374)2) 206.6 - 7,137.0

Calculation of column elongation due to hydraulic load (Equation 23):

TRL

E '" 9.54·10~----

Rl DDU2 _ DDl'l

EruA = 9.54 * 10-6 * 24,871.1 * 219/ «6.626)2 - (6.067)2)

Calculation of column shortening due to inside pressure (Equation 24): 1.37·1O-s(2P D - LPr '10-3)L

ER2 '" ---___;;~---'---

(DDU/DDI)'l -1

E]U"t = 1.37 * 10-5 (2 * 206.6 - 219 * 935 * 10-3) 219 / [(6.626/6.067)2 - 1]

Net column elongation (Equation 25):

EJt! = 7.3 - 3.2

Relative movement of the impellers in the bowl unit (Equation 26):

E£ '" EA-ER

Eg4 = 10.6 - 4.1

= 7,137.0 N

= lO.6mm

= 24,871.1 N

= 73 m.m

= 32m.m

= 4.1 mm

= 6.5 mm

(19)

(20)

(22)

(23)

(24)

(25)

(26)

38

The relative movement of impellers must not exceed the bowl unit's axial clearance by some safety margin in order to guarantee problem-free operation. For &JKH 2900 rpm the axial clearance is 19 mm, which is more than enough in this case. It is, therefore, easy to adjust its impellers, when thermal equilibrium is reached. We suggest adjustment of impellers 13.5 mm up.

IV - Selection of suitable motor for the pump

Calculation of total weight of pump shaft and impellers (Equation 27):

Ts '" 41.4D~2 L + T/

(27)

Ts = 41.4 * (1.187)2 * 219 + (5.5 * 12/0.225)

= 13,067.9 N

Calculation of total thrust load on motor (Equation 28):

(28)

T y4. = 7,137.0 + 13,067.9

= 20,205_0 N

Required size of the motor (Equation 29 and 30-32):

(29)

Calculation of power required by pump (Eqution 30):

From the performance curve we get the efficiency of the pump:

f10A = 77 % f1vfJ = 78 % f1DC = 79 %

(30)

Adl = 1.315 * 10-5 ;I: 206.6 * 935 * 44.2/0.77

= 145.8BHP

Calculation of thrust bearing power loss (Equation 31):

AL = nTT'1.684'IQ-S

ALA = 2,900 * 20,205.0 '" 1.684 '" 10-8 Calculation of lineshaft power loss (Equation 32):

From mechanical friction in lines hafts , Table 10 gives f = 7.97 HP/L = 100 m. Inserting in Equation 32 gives:

(31)

= 1.0 lIP

(32)

A"" = 7.97 * 219 * 10-2

= 17.5 lIP

39

Total power required of motor.

Ay4 = 145.8 + 1.0 + 17.5

= 1643 BlIP

The suitable power of the motor for a pump assembly is at the highest load required in the three cases A, Band C, and according to Table 13 this is 164.7 BlIP. This value is correlated with pump performance curves. The motors are generally produced in series, i.e. 150, 175,200 BHP, therefore, we select the motor size 200 BHP for this pump with an approximate 20% safety margin.

Why is the motor selected with a 20% safety margin?

- To account for inaccuracy in the calculations. If we select the motor to be 175 BHP, then the safety margin is only 6%. Normally one should have a margin of 15% or more.

- Because the pump can overload the motor when producing at high flow rate, for example if wellhead pressure (PH) becomes 0 bar.

- In the future, it may be found necessary to increase the number of impellers (stages), because the well drawdown will probably increase.

TABLE 13: Summation of calculated parameters

VARIATION IN WATER TABLE KV2
CALCULATED PARAMETERS 60m 80m 100 m
Water drawdown KN (m) 65.8 60.6 55.5
Submergence length hmin (m) 27.5 27.5 27.5
Minimum length of column Linin (m) 179.3 194.1 209.0
Flow rate Q (Vs) 44.2 424 40.6
Hydraulic pressure head loss Pd (m) 0.41 0.38 0.35
Column pressure bead loss Pp (m) 20.1 185 17.0
Total dynamic pressure head PD (m) 206.6 218.7 230.8
Hydraulic thrust on shaft TA (N) 7,137.0 7,555.0 7,973.0
Dynamic thrust on column T R (N) 24,871.1 2Jj,327.7 27,784.4
Weight of shaft and impellers TS (N) 13,067.9 13,067.9 13,067.9
Total thrust on motor T T (N) 20,205.0 20,623.0 21,041.0
I Elongation of shaft EA (mill) 10.6 11.2 11.8
Elongation of column due to dynamic load ERl (111m) 8.2 7.8 8.2
Shortening of column due to pressure Em (mm) 3.2 3.6 4.0
Net column elongation ER (mm) 4.1 4.2 4.2
Relative movement of impeller EE (mm) 6.5 7.0 7.6
Efficiency of pump TIn (%) 77 78 79
Power required by pump An (BHP) 145.8 146.2 145.8
Power loss on thrust bearing A L (HP) 1.0 1.0 1.0
Power loss of lineshaft A A (HP) 17.5 17.5 17.5
Total power required of motor AM (BHP) 1643 164.7 1643 v - Verify the limitation of the equipment

It is very important that the relative movement of impellers (EE) does not exceed the maximum lateral clearance of the pump. If the relative movement of the impellers is larger than the clearance, one must apply non-standard measures and find out by recalculation (III) if it is suffificient. In our case, the maximum calculated EE conclusion is 7.6 mm (Table 13), but the

40

maximum axial clearance of the pump allows 19 mID, so the &JKH 2900 rpm with the assemblies above is a suitable pump for this well.

The shaft horsepower ratings are given in Table 11. For shaft diameter 1 3f1&' at 2900 rpm, maximum horsepower is 184 BHP and maximum thrust 5,000 Ibs (22,222 N). The calculations show that the total thrust load on the shaft, T l' is 21,041 N (Table 13) which is inside the maximum limit, but as the horsepower is 9% higher there is a reson to be carefuL

The maximum thrust capacity of the 200 BHP motor is provided to 7,000 Ibs (31,100 N) (General Electric Motor 3292C). This is sufficient, because the maximum thrust on shaft is 21,041 N.

VI • Total weight of pump assembly

The maximum lifting capacity for the mobile crane is defined by Figure 14 and Table 7.

Total column length Column length per unit Number of column

Weight of column 6" x 3.042 m Column coupling

Cen tralizer

Enclosing tube

Bearing

Shaft

Shaft coupling

Total weight of string (not including bowl unit)

= 219 m = 3.042 m

= 72 unit

= 85.54 kg = 6.1 kg = 0.98 kg

:= 25.62 kg = 1.32 kg l: 17.71 kg = 0.45 kg

= 137.7 kg = 9,915 kg

Total weight per unit

41

5. INSTAlLATION OF GEOTIIERMAL DEEP WELL LINESIIAFf PUMPS

5.1 General

Proper installation is necessary to obtain maximum service life from the pump. To insure proper alignment three items are very important during installation:

1. All machined mating surfaces must be clean and free of burrs and nicks removed with a fine file or wire brush.

2. Exterior strain must not be transmitted to the pump discharge head. The most common cause of trouble in this respect is forcing the piping to mate with the pump. It is recommended that a flexible connector is installed in the piping adjacent to the pump.

3. All threads should be checked for damage and repaired if necessary. Clean all threads with wire brush and cleaning solvent, end of shafts must be cleaned and any burrs removed since alignment depends on the shaft ends butting squarely. Lubricate all screwed connections with a thread lubricant; a mixture of graphite grease is suitable for steel, an anti galling compound such as "Never-Seez" should be used on stainless and monel mating threads.

5.2 Installation

Preliminary:

a. Center lifting device over well opening and completion assembly.

b. Arrange two heavy timbers or l-beams across foundation to rest pipe clamps on.

c. Check shaft end play of bowl assembly by measuring distance shaft travels when pulled up then pushed all the way down. End play must be equal to or greater than total elongation of the unit indicated by Figure 21 and previous calculations.

Installation of bowl assembly:

a. Attach second pipe clamp near top of bowl assembly, lift bowl assembly and center over suction pipe.

b. Lower bowl assembly and screw onto suction pipe.

c. Lower bowl assembly until clamp rests on timbers, check enclosing tube and shaft projection (Figure 22).

Installing threaded column assembly:

a. Lineshafts are carefully straightened and placed inside the enclosing tube assembly for shipping. Each shaft should be checked by rotating several times to make sure it is not binding. Make sure each column pipe coupling is securely assembled to the column pipe.

b. Tie a rope to the lower end of the column pipe using a timber or clove hitch, then take a half hitch around the tube, and a reverse double half hitch around the lineshaft at the threads (Figure 23).

c. Attach pipe clamp near top of column pipe below coupling and raise the assembly into position over bowl assembly.

d. Slowly lower column section until lineshaft can be connected to the bowl shaft Place a small amount of lubricant on the shaft threads and screw lineshaft into coupling. Threads are LEFT HAND. Make sure the shafts. are butted together but do not use excess force in tightening. Wrenches should not be needed except for the final fraction of a tum to

30,000
20,000
15,000
10,000
11'1 8,000
C 7,000
Z
::I 6,000
~ 5,000
!Ii 4,000
~
11'1
:) 3,000
II:
:
~
u 2,000
:l
::I
< 1,500
II:
0
>
:c 1,000
800
700
600
500
400
300
200
150
100 42

EXAMtLE: // V/ V ~ ~ ~ V V V /
/ /. /
Hydra~lic ~r11 = 3,SOOlbs. ~ Vj i~ ~ ~ V V / V
Shall'" 1 1/2" • ~ / V
/,<'
S811inv .. 400 f~ • ijJ ",'\'Y~ ~~ I"V V V / V
/
Total Elongation '" .33 ,b ~/V~~ / V /
,,!>':>\,.._'~ ,'" ,\.( V /
)'!- S.. I.'~ // " /
/ 1/ / '\. "-',b '" / / /
1/ 1/ 1/ 1/ V ~'o 1/ 1/ "
Vl/ / r1' -, ", / V· V
/ ,IV ; /// .s: / V
1/// V ~'f"'// / , V /
IV
f4.- 1-/ // /~ ~ ~ ./ / "V
V 1:9' / V",\
// :% V ~ i~ ~ V V V /
/ ~ V /, /
V/ 'l f"~ VI' V / V Q')/
./ ",""
~ ~ ~ ~ vV /1 ... ~
.. /
if // 1/
// '/ / /
¥ // / / ,/ /
1/1/ V I'
V/ / / V
V / V I'
/ / V I
/
/ ./V I
I ,
/ ~ ~ )// IV / V /VL V / L_
V
/ I 0 ~ 'i 1/ / / IV V / 1/
/1 l/
I /. / /
/ L l/
'// / / /
/j /1/ 1/ 1/ 1/ 1/
/1/ /1/ / 1/ / / /
/V/~ // / / / / /
.. ~ /~ v/ I' / / /
V. /
vt;: ~ ~ ~ ~/ ./ ~/ ~ l/.t- V-..a L. I- ... -
- ~ F- ~
// vv tv / / 1/ / / L /
Vh v/ / / / / /v / V
v; :/V~ V Vv / v/ V / / V
Ev ~ vV I v // V V Vv v·
/ '/ V / /
T r I I I I I 1000 1100 600 400 300 200 150 1 00 75 50 SETTING IN FEET

FIGURE 21: Shaft elongation (Floway, 1970)

2.00
1.50 11'1
11.1
:
u
1.00 !:
.so !:
.70 Z
.60 0
~
.50 <
CI
.40 Z
0
oJ
.30 11.1
.J
.25 <
I-
.20 0
I-
.15
.10 make the joint tight. Do not allow filings to drop in to the bearings.

e. Lower the enclosing tube and column pipe until the enclosing tube can be connected to the tube adaptor bearing. Lubricate tube adaptor bearing threads and screw enclosing tube on. Threads are LEFr HAND. Tighten with pipe wrench or small chain tong.

f. Lower column pipe into column coupling after lubricating threads. Threads are RIGHT HAND. Tighten with pipe wrench.

I.T=l JHO HSI:> 9000 UK LJ=J 91,10,0684 T

Lineshaft coupling

Lines haft

Lineshaft bearing

Pipe coupling

Column pipe

43

I,T:l JHD HSp 9000 UK L..tJ 91.10,0683 T

16"

FIGURE 22: Standard enclosing tube and lineshaft projection (Peabody Floway)

FIGURE 23: Installation of column assembly (Peabody Floway)

g. Lower assembly unit and clamp at top of column pipe. Lower assembly until it rests on timbers.

h. Install enclosing tube stabilizers at every column joint.

i. Repeat this procedure for the remaining sections except the top section with the following precautions. Keep all parts clean, check all threads, repairing any damage, do not bend lineshaft, make sure all joints butt and are tight, and check tube and shaft projection after each joint is installed.

Installing top column section:

a. The components of the top column section should be laid out for inspection and measuring. The top column section will consist of:

i. The top column pipe may be a standard 1.5-3 m length or special length required; this length of pipe will not have a coupling attached.

u. The tube adjusting nipple assembly is a special length of enclosing tube with one end turned smooth on the outside and extra long threads on the inside of the same end.

m, The top lineshaft is usually a special length, but is otherwise a standard shaft.

iv, The head shaft, extra long threads with key wayan top end for connecting to driver.

b. Measure the shaft and enclosing tube projection before the top column section is installed as illustrated in Figure 22. The enclosing tube projection must be between 7 1/2 and 9", while the shaft projection must be between 14 and 18".

44

~ JHD HSp 9000 UK L.U 91 10.0685 T

Headshaft nut

TYPEH&HN

_,_- Headshaft __ -- Shaft coupling /. Water lube assembly ~ • Adjustable flange coupling \ ~--Top lineshaft --~

L_~-!_-tr---Tension nut -----.1J . __ "'-+'~_j

TYPEF

~i::L-l:~::_-- Tension plate ---__'::::~~~~i Di1m2~>nrl ___ ____,..--- Tube adjusting nipple -~

~----Discharge head -------11

'Lf=~--.J:l

Top column adaptor flange

Column adjusting nipple Top column adaptor

Top column --------11. __ --....

FIGURE 24: Discharge head assemblies (Peabody Floway, 1984)

c. If a separate top lineshaft and head shaft are used, determine where the shaft coupling will be in relation to the top of tbe tube adjusting nipple. If the shaft coupling is located below the top of the tube adjusting nipple, the headshaft should be connected before proceeding with assembly. If the shaft coupling will be above the top of the tube adjusting nipple, the headshaft should be set aside for installation after the discharge head is installed.

d. Inspect the turned section of the tube adjusting nipple for dirt, rust, nicks or scratches.

Clean and polish with fine emery cloth as necessary to provide a smooth surface. Proceed with assembly by the same methods used for previous lengths and clean the "D-ring" groove.

Installing discharge bead:

a. The discharge head assembly will consist of the following major parts (Figure 24):

I. The discharge head may have either a flanged or threaded connection for

attaching the column pipe to the bottom of the head.

II. A tension plate is used for centering and sealing the enclosing tube.

Ill. A tension nut used to support the enclosing tube, provides the top bearing.

iv. The headshaft assembly connects the lineshaft to the driver.

b. A discharge head using screwed top column flanges are suggested for easier installation.

c. Lubricate tension nut threads, then lower into place and screw in tube adjusting nipple until tension nut comes down snug against tension plate. Continue to tighten tension nut until all of the enclosing tube is supported by the discharge head and is in slight tension. Proper tension is achieved by noting a gradual increase in the force needed to tighten the tension nut; when proper tension is reached, there will be a sudden increase in the

~ JHDHSPOOOO UK L.:J:.J 911 0 0600 T

45

water line

r:-r::l JHD H$p 9000 UK L__t_j 91 100687T

Discharge head Tube nipple "O" ring Tube adjusting nipple

Packing gland

Packing rings Injection type packing box: nut

lockscrew

Stuffing box Clutch

·0" ring

Tension plate

Discharge head

bearing

Headshaft nut lockscrew

Gib key

FIGURE 25: Water lubrication assembly (Peabody Floway)

FIGURE 26: Hollow shaft driver clutch (Peabody Floway)

tightening force required. The approximate amount that the enclosing tube must be raised must be approximately = 0.03 (feet of column assembly)2 10-3 (inches)

Installing water lubrication systems:

The installation of units using water injection lubrication system (Figure 25) in geothermal wells use filtered fresh hot water for injection lube at 02-03 Us (3-5 gpm) at 25-3.5 bar (40-50 psig) for lubricating the lineshaft bearings. Note that the by-pass ports of bowl unit are open.

Installing hollow shaft driver.

The electric driver (Figure 26) should be checked for rotation before installation. The driver must be rotated counter-clockwise, when looking down at the top end of the motor. To change the direction of rotation on a three phase motor, interchange any two line leads. Raise and center the driver over the pump carefully until it sits on the discharge head and then tighten the mounting bolts. Check that the alignment of the shaft will center in the shaft clearance space at the top of the driver. If the shaft is not centered, place shims or leveling wedges between the discharge head and the foundation unti1 the shaft is centered.

Impeller adjustment hollow shaft driver.

When using a hollow shaft driver impeller adjustment is accomplished at the top of the driver by the following procedure. The driver canopy will have to be removed before beginning. The headshaft nut screws down (Right hand threads), until impeller are just lifted off their seat and the shaft will rotate freely. When semi-open impellers are used, the correct determination of the point where the impellers just barely clear their seat is very important for proper adjustment. Adjust impellers as calculated before and according to shaft elongation graph. Lock heads haft nut withlockscrew inserted down through hole in headshaft nut and threaded into driver clutch.

46

Discharge piping:

The discharge piping must be installed so that stresses are not transmitted to the pump. A fixible connector installed next to the discharge head is highly recommended. Do not use the discharge head to support the piping and fittings, and support with blocks or saddles.

r:T'=1 JHO HSp 0000 UK Ll=J 91 10.01388 T

Capillary tube installation and use: Tire

The capillary tube installed with the pump (Figure 27) is a simple and effective method of determining the depth of water

in the well. Steel, stainless steel or plastic tubing can be use for the capillary tube.

The lower end of tube is near the bottom

of the pump. For reliable reading, the capillary tube should extend 7 m below lowest water level if possible. The upper

end of the capillary tube is connected to a gauge and snifter valve. The exact vertical length of the capillary tube must be recorded (L ).. A tire pump or air compressor is used to expel all water from

the tube; when this point is reached the gauge reading will remain constant. The maximum maintained pressure is equal to

the height of the water above the end of

the capillary tube (D). The water level (below the surface) is obtained by subtracting "D" from "L" (lVL = L - D or

WI. = L -Ppg).

Gauge

Water level

Approx.2'

f

FIGURE 27: Air line installation and use (Peabody Floway)

47

6. CONCLUSIONS

1. The study describes three methods of pump testing a geothermal well, the procedure to be followed and the data collection required.

2. Interpretation of pump test results is described and a simplified analytical model used for pump selection.

3. A systematic procedure (flow diagram) for pump design and selection is presented.

4. Engineering equations used in detailed pump design are identified and explained. which will enable their use for other conditions than for the example used in this study.

5. The influence of variable water tables on pump output is analyzed and a pump selected accordingly.

6. The use of hydraulic balanced impellers is demonstrated for reducing shaft elongation.

7. Impeller adjustment calculations are made taking into account the different forces affecting shaft and column elongation.

8. The material selection for geothermal lineshaft pumps is described.

9. Proper procedures for installing and pulling a deep well pump are given. Critical items are identified such as shaft straightness and bearing clearances.

48

ACKNOWLEDGEMENTS

I would like to express my gratitude to the respected staff of the UNU and aU persons of Orkustofnun, for their guidance and useful advice, thus making possible the successful completion of the training programme.

Special thanks to: Dr. Ingvar B. Fridleifsson as the Director of the United Nations University Geothermal Training Programme in Iceland, who gave me the honour of being the fifth participant from Thailand to receive training, and for his continuous follow up and care of aU UNU fellows; Sverrir Thorhallsson for being an excellent advisor who gave me knowledge and continuous supervision, discussing many things until this training finalized.

I also mention with gratitude: Ami Gunnarsson consulting engineer for his lectures on design and selection of deep well pumps; Ari Rognvaldsson and Snaelaugur Stefansson from Hitaveita Akureyrar for practice in the installation and repairing of a deep well pump; Omar Sigurdsson for well testing analysis; and the reservoir engineering staff for special course lectures.

I extend my thanks to Ludvik S. Georgsson and Marcia Kjartansson for their kind care for UNU fellows from the first day to the last of training.

Finally, my thanks to the UNU fellows for their understanding and companionship during my training in Iceland.

49

NOMENCIATURE

A = Initial water level m
AA = Power loss in lineshaft HP
AD = Power required of pump BHP
AL = Power loss in motor thrust bearings HP
AM = Power required by motor = AA + AL + AD BHP
AN = Power consumption of motor HP
B = Laminar drawdown coefficient m/(I/s)
C = Turbulent drawdown coefficient m/(Vs)2
C1 = Well drawdown coefficient m/(Vsi
C2 = Coefficient of head loss in column m/(l/si
c = Compressibility Pa-1
DDI = Inner diameter of column in
DDU = Outer diameter of column In
Do = Diameter of pump shaft in
DSI = Inner diameter of enclosing tube In
DSU = Outer diameter of enclosing tube in
E = Voltage per leg applied to motor volt
EA = Elongation of shaft due to hydraulic thrust mm
EE = Relative movement of impeller mm
ER = Total elongation of column mm
Em = Elongation of column due to total pressure head mm
ER2 = Shortening of column due to inside pressure mm
f = Coefficient of mechanical friction in lineshaft bearings HP/L=l00 m
g = Acceleration of gravity mls2
H = Water level in the well m
1JH = Well drawdown m
h = Reservoir thickness m
h = Pumping submergence m
hmin = Minimum submergence of pump in water m
I = Amperes per leg applied to motor ampere
K = Thrust constant lhlft
KD = Portion of pump below water level m
KN = Well drawdown m
Kv = Total static water level m
Kvz = Elevation above sea level to well head m a.s.!.
KV] = Elevation below sea level to static water level m b.s.l,
k = Permeability of the reservoir m2
L = Pump setting level = Kv + KN + KD m
Lmin = Minimum length required of column m
n = Rotation of motor rpm
NPSHR = Net positive suction head required m
p. = Pressure minimum required bar
mUI
Pa = Vaporize pressure of water bar
PD = Total pressure head m
Pd = Dynamic pressure in column m
PF = Pressure loss in column m
PI = Power factor of motor
PH = Discharge head pressure m
Pu = Atmospheric pressure bar Q QB Qt

'w '1 S

T TA T[

TR TS TT t

V Yll Y21 Y12 Y22 Z

Greek letters:

50

= Well discharge

== Bast point efficiency of discharge

== Discharge from or injection into the well being tested == Radius of discharge well

= Distance between the discharge and observation well = Storativity

:= Transmissivity

= Hydraulic thrust on pump shaft = Weight of impeller

= Hydraulic thrust on column = Weight of pump shaft

= Total thrust on pump shaft = time, since well test started == Velocity of water in column

= Coefficient of pump curve (Q < QB) == Coefficient of pump curve (Q < QB) = Coefficient of pump curve (Q ~ QB) :: Coefficient of pump curve (Q ~ QB) = Number of pump impeller

= Efficiency of motor = Efficiency of pump

== Dynamic viscosity of the reservoir fluid = Density of fluid

= Density of fluid at T'C

= Porosity of the reservoir rocks

IJs

IJs

lis

m

m mJPa m3tpa.s N

N

N

N

N

sec

m/s m/(IJs)

m

m/(IJs)

m stage

%

% Pa.s kg/m3 kg/m3

51

REFERENCES

British Standards Institution, 1983: Test pumping water wells. BS 6316.

Driscoll, F.G., 1986: Groundwater and wells. 2nd edition, Johnson Division, St Paul, Minn., U.S.A, 1089 pp.

Floway, 1970: Technical information sheets. Floway pumps, Fresno, California, U.S.A

Kjaran, S.P., Eliasson, J., 1983: Geothermal reservoir engineering lecture notes. UNU G.T.P., Iceland, report 2, 250 pp.

Peabody F1oway: Well pumps, enclosed lineshaft types. Instruction manuals, no. IM-I02 and IM- 107B, Fresno, California, U.S.A

Peabody Floway, 1988: Vertical Turbine Pumps. Turbine Data Handbook. 2nd edition, Peabody Floway, Fresno, California, U.S.A

Rafferty, K., and Culver, G., 1989: Well pumps. OIT Geo-Heat Center, Klamath Falls, Oregon, U.S.A

Todd, D.K., 1980: Groundwater Hydrology. John Wiley & Sons, New York, U.S.A, 535 pp.

Design and Select Pumps

Uploaded by

Copyright:

Available Formats

You might also like

Design and Select Pumps

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Design and Select Pumps

Uploaded by

Copyright:

Available Formats

Report 8, 1991

DESIGN AND SEIECITON OF DEEP

You might also like

Design and Select Pumps

Uploaded by

Copyright:

Available Formats

You might also like

Design and Select Pumps

Uploaded by

Document Information

Original Description:

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Design and Select Pumps

Uploaded by

Copyright:

Available Formats

﻿Report 8, 1991

DESIGN AND SEIECITON OF DEEP

You might also like

Report 8, 1991